Pre-Clinical Investigation of a Novel Small Molecule Inhibitor Targeting YBX1 in AML

Recently, our group identified the pleiotropic DNA and RNA binding protein YBX1 as a specific dependency and therapeutic target in acute myeloid leukemia (AML) that is essential for stabilizing oncogenic transcripts and thereby maintains oncogenic protein expression (Perner et al., 2021). Selective pharmacologic targeting of YBX1 in AML, however, has not been achieved so far. Here, we investigate the functional consequences of a novel small molecule inhibitor (SU056) of YBX1 in AML.

We examined the viability of several human AML cell lines (MOLM13, HL60, EOL1, THP1, Kasumi1, ML2, Mv4;11, OCI-AML3, NOMO1) upon SU056 treatment and found significant reduction of cell growth in a concentration dependent manner. Inhibition of cell growth could be mainly attributed to impaired proliferative capacity at nanomolar concentrations. Pharmacologic inhibition of YBX1 resulted in impaired colony formation (mean colony number DMSO: 231.3±13.88 vs. SU056 500nM: 154.7±8.29; p<0.0001), cell cycle arrest (cells in S phase: DMSO: 37.3±2.4 vs. SU056 500nM: 27.2±1.67; p=0.0127) and induction of myeloid differentiation. Consistently, in a xenograft model of human AML, SU056 treatment resulted in prolonged disease latency (MOLM13 diluent: 62 days OS vs. SU056: not reached, p=0.0013; OCI-AML3 diluent: 79 days OS vs. not reached, p=0.0001) and significant reduction of disease penetrance (MOLM13 diluent: 100% vs. SU056: 24,9%; OCI-AML3 diluent: 100% vs. SU056: 9,1%). Consistently, percentage of SU056 treated human leukemic cells in the bone marrow of recipient mice was decreased for both, MOLM13 (mean diluent: 41,3% vs. SU056: 7,6%) and OCI-AML3 (mean diluent: 28,9% vs. SU056: 2,9%). Likewise, pharmacologic inhibition of YBX1 reduced colony formation and total cell numbers of primary human AML cells in a dose-dependent manner.

To be clinically relevant, such treatment must have a suitable therapeutic index between human hematopoietic stem- and progenitor cells (HSPCs) and their malignant counterpart. This is particularly important, considering the role of YBX1 in mRNA splicing and processing. Of note, pharmacologic inhibition of YBX1 did not perturb colony formation of normal HSPCs derived from apheresis donors.

These findings were confirmed by investigating primary human leukemic cells in a patient-derived xenograft (PDX) model of AML. We aimed to assess whether loss in leukemia cell competition induced by pharmacologic targeting of YBX1 is sufficient for normal hematopoietic cells to outcompete leukemic cells in vivo. 1x10⁶ normal human HSPCs were transplanted along with 2x10⁴ leukemic cells to assess for efficacy of SU056-treatment versus diluent control. In vivo treatment with SU056 (20mg/kg, i.p.) for 3 weeks resulted in a significant decrease in the AML blast number in the PB (mean blast %: diluent: 38.5%; SU056: 2.25%; p<0.0001). This resulted in increased survival of SU056 treated recipients (OS diluent: 83 days vs. SU056: not reached).

To confirm the proposed mechanism by which SU056 impairs leukemia growth we performed global transcriptome and proteome analysis following treatment with SU056 (500nM, 96h). Proteome profiling revealed a total of 2641 deregulated proteins. Gene-ontology analysis indicated deregulation of YBX1 targets and proteins associated with cell cycle, DNA repair, mRNA splicing/processing, and transcriptional regulation. Of note, overlap between transcriptional regulation (as determined by mRNA-Seq) and changes in proteome abundance revealed less than 5% target genes/proteins, indicating that SU056 may in fact interfere with binding of oncogenic mRNAs to YBX1 and thereby may reduce translation of oncogenic transcripts. In order to define functional consequences of pharmacologic YBX1 inhibition and explore potential vulnerabilities of drug combinations, we performed a genome-wide CRISPR/Cas9 screen in human AML cells treated with SU056 versus diluent control. Validation experiments are currently ongoing.

Taken together, our data confirm the therapeutic potential to target YBX1 using the novel small molecule SU056. Given its broad applicability across genetic subtypes of AML and lack of toxicity on normal HSPCs, SU056 as a single agent or in combination with established targeted therapies may efficiently help to target and outcompete leukemic cells. Experiments to assess for potential synergy with already established targeted therapies will be presented.

No relevant conflicts of interest to declare.